Forebrain
# splicing <- c("SE","MXE","RI","A3SS","A5SS")
# splice_type <- c()
# mouse <- read.delim("~/Desktop/MISO_proj/EuropeanDataset/data/Sample_Metadata_Mouse_Heart-Brain-Only.tsv",sep="\t",header = T,stringsAsFactors = F)
# mouse <- mouse[order(mouse$Source.Name),]
fore.samps <- mouse$Source.Name[which(mouse$Characteristics.organism.part.=="forebrain")]
# Combine the matrices into one matrix
# for (splice in splicing){
# x <- read.csv(paste("/Users/alisha/Desktop/faster_Datastore/Alisha/MISO_project/EuropeanDataset/mouse/cohort_level/miso_counts_",splice,"_mat.csv",sep=""),header=T,row.names = 1,check.names = F)
# colnames(x) <- sub(".counts.txt","",colnames(x))
# x <- x[,order(colnames(x))]
# y <- apply(x,1,function(row){table(is.na(row))["FALSE"] >= half})
# x <- x[y==T,]
# splice_type <- c(splice_type,rep(splice,nrow(x)))
# all_psi <- rbind(all_psi,data.frame(x,SpliceType=splice))
#
# }
fbrain <- data.frame(Samples = mouse$Source.Name[which(mouse$Characteristics.organism.part.=="forebrain")], DevelopStage = mouse$Characteristics.age.[which(mouse$Characteristics.organism.part.=="forebrain")])
fbrain$Group <- ""
fbrain$Group[which(fbrain$DevelopStage == 28.0 | fbrain$DevelopStage == 63.0)] <- "Adult"
fbrain$Group[which(fbrain$DevelopStage == 14.0)] <- "Adolescent"
fbrain$Group[which(fbrain$DevelopStage == 3.0 | fbrain$DevelopStage == 0.0)] <- "Postnatal"
fbrain$Group[which(fbrain$DevelopStage == 17.5 | fbrain$DevelopStage == 18.5)] <- "Very_Late_Embryo"
fbrain$Group[which(fbrain$DevelopStage == 15.5 | fbrain$DevelopStage == 16.5)] <- "Late_Embryo"
fbrain$Group[which(fbrain$DevelopStage == 13.5 | fbrain$DevelopStage == 14.5)] <- "Midstage_Embryo"
fbrain.psi <- all_psi[,fore.samps]
fbrain.psi.complete <- fbrain.psi[complete.cases(fbrain.psi),]
fbrain.psi.complete <- fbrain.psi.complete[,order(colnames(fbrain.psi.complete))]
fbrain.psi.100 <- fbrain.psi.complete * 100 # turn psi values into integers
fbrain.psi.100 <- round(fbrain.psi.100,0)
design <- model.matrix(~fbrain$Group)
y = DGEList(counts=fbrain.psi.100, group = fbrain$Group)
y <- estimateDisp(y,design)
fit <- glmQLFit(y, design)
fit <- glmQLFTest(fit, coef=2:6)
tab <- as.data.frame(topTags(fit, n=450))
tab$Gene <- sapply(rownames(tab),function(name){translate$GeneSymbol[which(translate$ID==name)]})
# write.table(tab, file = "/Users/alisha/Desktop/MISO_proj/EuropeanDataset/Mouse_Forebrain_DEisoforms.tsv",sep="\t",col.names = NA)
# kable(head(tab)) %>% kable_styling()
# translate <- as.data.frame(fread("/Users/alisha/Desktop/faster_Datastore/Alisha/MISO_project/data/mouse/allSplice.mm10lnc.gff3.truncated.tsv"))
top5 <- c("chr9:41579978:41580198:+@chr9:41580713:41581402:+@chr9:41590220:41592829:+",
"chr12:109637839:109637926:+@chr12:109640267:109640328:+@chr12:109642565:109642652:+",
"chr9:41528123:41528200:+@chr9:41576408:41576549:+@chr9:41589974:41590167:+",
"chr14:55056024:55056177|55056430:+@chr14:55097380:55098986:+",
"chr12:109637839:109637926:+@chr12:109640267:109640328:+@chr12:109642565:109642652:+@chr12:109643414:109643461:+")
top5 <- as.data.frame(fbrain.psi.complete[top5,])
top5 <- merge(top5,translate,by.x=0,by.y=3, all.x = T, all.y=F)
fbrain$Group <- factor(fbrain$Group, levels=c("Midstage_Embryo",
"Late_Embryo",
"Very_Late_Embryo",
"Postnatal",
"Adolescent",
"Adult"))
top5.melt <- reshape2::melt(top5)
top5.melt <- merge(top5.melt,fbrain,by.x=4,by.y = 1)
Top 5 Most Significantly DE Isoforms
ggplot(top5.melt,aes(x=Group, y=value,fill=Group)) + geom_boxplot() + ylab("PSI") +
facet_wrap(~Row.names) + theme_bw() +
theme(axis.text.x = element_text(color = "black",size = 14, angle = 90, hjust = 1,vjust = 0.5),
strip.text.x = element_text(color="black",face = "bold",size=7))
new <- fbrain.psi.complete[rownames(tab)[which(tab$FDR<0.05)],]
rownames(new) <- sapply(rownames(new),function(name){paste(translate$GeneSymbol[which(translate$ID==name)],name,sep="_")})
new$transcripts <- rownames(new)
new.melt <- melt(new)
new.melt <- merge(new.melt,fbrain,by.x=2,by.y=1)
# head(new.melt)
transcripts <- unique(new.melt$transcripts)
pdf("/Users/alisha/Desktop/MISO_proj/EuropeanDataset/Mouse_Forebrain_DEisoforms_Boxplots.pdf",width = 9,height = 7)
for (trans in transcripts){
df <- new.melt[which(new.melt$transcripts==trans),]
print(ggplot(df,aes(x=Group, y=value,fill=Group)) + geom_boxplot() + ylab("PSI") + theme_bw() +
ggtitle(paste(str_wrap(trans, width = 75))) + xlab("") +
theme(axis.text.x = element_text(color = "black",size = 14, angle = 90, hjust = 1,vjust = 0.5),
plot.title = element_text(hjust=0.5,size = 12), legend.position = "none") )
}
dev.off()
## quartz_off_screen
## 2
fbrain <- fbrain[order(fbrain$Group),]
ra <- HeatmapAnnotation(Group = fbrain$Group,col = list(Group=c("Postnatal"= "purple",
"Adult" = "coral3",
"Midstage_Embryo"="blue",
# "Early_Embryo" = "yellow",
"Late_Embryo" = "brown",
"Very_Late_Embryo"="black",
"Adolescent"= "cyan3")))
# "Developing_Embryo"="deeppink1")))
#ha <- HeatmapAnnotation(SpliceType = splice_type, col=list(SpliceType=c("SE"="red","MXE"="blue","RI"="cyan3","A3SS"="yellow","A5SS"="orange")))
lt05 <- rownames(tab)[which(tab$FDR < 0.05)]
toplot <- fbrain.psi.complete[lt05,]
toplot <- medianCtr(toplot) #log2(toplot+1))
draw(Heatmap(as.matrix(toplot), cluster_columns = F, name = "PSI",column_title = "Top DE Transcripts (FDR < 0.05)\nMouse Forebrain Samples", clustering_distance_columns = "pearson",clustering_method_columns = "average", bottom_annotation = ra, row_names_max_width = max_text_width(rownames(toplot), gp = gpar(fontsize = 3))),heatmap_legend_side = "left", annotation_legend_side = "left")
kable(fbrain) %>% kable_styling()
|
|
Samples
|
DevelopStage
|
Group
|
|
28
|
2223sTS.Mouse.Brain.14.5.Male
|
14.5
|
Midstage_Embryo
|
|
29
|
2227sTS.Mouse.Brain.14.5.Female
|
14.5
|
Midstage_Embryo
|
|
30
|
2237sTS.Mouse.Brain.14.5.Male
|
14.5
|
Midstage_Embryo
|
|
31
|
2243sTS.Mouse.Brain.14.5.Female
|
14.5
|
Midstage_Embryo
|
|
41
|
2813sTS.Mouse.Brain.13.5.Female
|
13.5
|
Midstage_Embryo
|
|
42
|
2821sTS.Mouse.Brain.13.5.Male
|
13.5
|
Midstage_Embryo
|
|
43
|
2829sTS.Mouse.Brain.13.5.Female
|
13.5
|
Midstage_Embryo
|
|
44
|
2837sTS.Mouse.Brain.13.5.Male
|
13.5
|
Midstage_Embryo
|
|
25
|
2185sTS.Mouse.Brain.15.5.Male
|
15.5
|
Late_Embryo
|
|
26
|
2189sTS.Mouse.Brain.15.5.Female
|
15.5
|
Late_Embryo
|
|
27
|
2193sTS.Mouse.Brain.15.5.Male
|
15.5
|
Late_Embryo
|
|
32
|
2247sTS.Mouse.Brain.15.5.Female
|
15.5
|
Late_Embryo
|
|
33
|
2254sTS.Mouse.Brain.16.5.Female
|
16.5
|
Late_Embryo
|
|
34
|
2257sTS.Mouse.Brain.16.5.Male
|
16.5
|
Late_Embryo
|
|
35
|
2273sTS.Mouse.Brain.16.5.Female
|
16.5
|
Late_Embryo
|
|
36
|
2276sTS.Mouse.Brain.16.5.Male
|
16.5
|
Late_Embryo
|
|
21
|
2157sTS.Mouse.Brain.17.5.Male
|
17.5
|
Very_Late_Embryo
|
|
22
|
2160sTS.Mouse.Brain.17.5.Male
|
17.5
|
Very_Late_Embryo
|
|
23
|
2163sTS.Mouse.Brain.17.5.Female
|
17.5
|
Very_Late_Embryo
|
|
24
|
2165sTS.Mouse.Brain.17.5.Female
|
17.5
|
Very_Late_Embryo
|
|
37
|
2307sTS.Mouse.Brain.18.5.Male
|
18.5
|
Very_Late_Embryo
|
|
38
|
2313sTS.Mouse.Brain.18.5.Female
|
18.5
|
Very_Late_Embryo
|
|
39
|
2317sTS.Mouse.Brain.18.5.Male
|
18.5
|
Very_Late_Embryo
|
|
40
|
2323sTS.Mouse.Brain.18.5.Female
|
18.5
|
Very_Late_Embryo
|
|
1
|
1740sTS.Mouse.Brain.0dpb.Female
|
0.0
|
Postnatal
|
|
8
|
1900sTS.Mouse.Brain.0dpb.Female
|
0.0
|
Postnatal
|
|
9
|
1906sTS.Mouse.Brain.0dpb.Male
|
0.0
|
Postnatal
|
|
10
|
1912sTS.Mouse.Brain.0dpb.Male
|
0.0
|
Postnatal
|
|
11
|
1918sTS.Mouse.Brain.3dpb.Female
|
3.0
|
Postnatal
|
|
12
|
1924sTS.Mouse.Brain.3dpb.Male
|
3.0
|
Postnatal
|
|
13
|
1930sTS.Mouse.Brain.3dpb.Female
|
3.0
|
Postnatal
|
|
20
|
1964sTS.Mouse.Brain.3dpb.Male
|
3.0
|
Postnatal
|
|
2
|
1876sTS.Mouse.Brain.2wpb.Male
|
14.0
|
Adolescent
|
|
3
|
1880sTS.Mouse.Brain.2wpb.Male
|
14.0
|
Adolescent
|
|
4
|
1884sTS.Mouse.Brain.2wpb.Female
|
14.0
|
Adolescent
|
|
5
|
1888sTS.Mouse.Brain.2wpb.Female
|
14.0
|
Adolescent
|
|
6
|
1892sTS.Mouse.Brain.4wpb.Male
|
28.0
|
Adult
|
|
7
|
1896sTS.Mouse.Brain.4wpb.Male
|
28.0
|
Adult
|
|
14
|
1936sTS.Mouse.Brain.4wpb.Female
|
28.0
|
Adult
|
|
15
|
1940sTS.Mouse.Brain.4wpb.Female
|
28.0
|
Adult
|
|
16
|
1944sTS.Mouse.Brain.9wpb.Male
|
63.0
|
Adult
|
|
17
|
1948sTS.Mouse.Brain.9wpb.Female
|
63.0
|
Adult
|
|
18
|
1954sTS.Mouse.Brain.9wpb.Male
|
63.0
|
Adult
|
|
19
|
1960sTS.Mouse.Brain.9wpb.Female
|
63.0
|
Adult
|
Hindbrain
splicing <- c("SE","MXE","RI","A3SS","A5SS")
hind.samps <- mouse$Source.Name[which(mouse$Characteristics.organism.part.=="hindbrain")]
hbrain <- data.frame(Samples = mouse$Source.Name[which(mouse$Characteristics.organism.part.=="hindbrain")], DevelopStage = mouse$Characteristics.age.[which(mouse$Characteristics.organism.part.=="hindbrain")])
hbrain$Group <- ""
hbrain$Group[which(hbrain$DevelopStage == 28.0 | hbrain$DevelopStage == 63.0)] <- "Adult"
hbrain$Group[which(hbrain$DevelopStage == 14.0)] <- "Adolescent"
hbrain$Group[which(hbrain$DevelopStage == 3.0 | hbrain$DevelopStage == 0.0)] <- "Postnatal"
hbrain$Group[which(hbrain$DevelopStage == 17.5 | hbrain$DevelopStage == 18.5)] <- "Very_Late_Embryo"
hbrain$Group[which(hbrain$DevelopStage == 15.5 | hbrain$DevelopStage == 16.5)] <- "Late_Embryo"
hbrain$Group[which(hbrain$DevelopStage == 13.5 | hbrain$DevelopStage == 14.5)] <- "Midstage_Embryo"
# colnames(all_psi) <- sub("^X","",colnames(all_psi))
hbrain.psi <- all_psi[,hind.samps]
hbrain.psi.complete <- hbrain.psi[complete.cases(hbrain.psi),]
hbrain.psi.complete <- hbrain.psi.complete[,order(colnames(hbrain.psi.complete))]
hbrain.psi.100 <- hbrain.psi.complete * 100 # turn psi values into integers
hbrain.psi.100 <- round(hbrain.psi.100,0)
design <- model.matrix(~hbrain$Group)
y = DGEList(counts=hbrain.psi.100, group = hbrain$Group)
y <- estimateDisp(y,design)
fit <- glmQLFit(y, design)
fit <- glmQLFTest(fit, coef=2:6)
tab <- as.data.frame(topTags(fit, n=500))
tab$Gene <- sapply(rownames(tab),function(name){translate$GeneSymbol[which(translate$ID==name)]})
write.table(tab, file = "/Users/alisha/Desktop/MISO_proj/EuropeanDataset/Mouse_Hindbrain_DEisoforms.tsv",sep="\t",col.names = NA)
# kable(head(tab)) %>% kable_styling()
top5 <- c("chr8:87524296-87524427:+@chr8:87525415-87525552:+",
"chr12:109627048:109627153:+@chr12:109634194:109634354:+@chr12:109635394:109635457:+",
"chr9:41579978:41580198:+@chr9:41580713:41581402:+@chr9:41590220:41592829:+",
"chr12:109742062:109742290:+@chr12:109743357:109743418:+@chr12:109743666:109743847:+",
"chr12:109637839:109637926:+@chr12:109640267:109640328:+@chr12:109642565:109642652:+")
top5 <- as.data.frame(hbrain.psi.complete[top5,])
top5 <- merge(top5,translate,by.x=0,by.y=3, all.x = T, all.y=F)
hbrain$Group <- factor(hbrain$Group, levels=c("Midstage_Embryo",
"Late_Embryo",
"Very_Late_Embryo",
"Postnatal",
"Adolescent",
"Adult"))
top5.melt <- reshape2::melt(top5)
top5.melt <- merge(top5.melt,hbrain,by.x=4,by.y = 1)
Top 5 Most Significantly DE Isoforms
ggplot(top5.melt,aes(x=Group, y=value,fill=Group)) + geom_boxplot() + ylab("PSI") +
facet_wrap(~Row.names) + theme_bw() +
theme(axis.text.x = element_text(color = "black",size = 14, angle = 90, hjust = 1,vjust = 0.5),
strip.text.x = element_text(color="black",face = "bold",size=7))
new <- hbrain.psi.complete[rownames(tab)[which(tab$FDR<0.05)],]
rownames(new) <- sapply(rownames(new),function(name){paste(translate$GeneSymbol[which(translate$ID==name)],name,sep="_")})
new$transcripts <- rownames(new)
new.melt <- melt(new)
new.melt <- merge(new.melt,hbrain,by.x=2,by.y=1)
# head(new.melt)
transcripts <- unique(new.melt$transcripts)
pdf("/Users/alisha/Desktop/MISO_proj/EuropeanDataset/Mouse_Hindbrain_DEisoforms_Boxplots.pdf",width = 9,height = 7)
for (trans in transcripts){
df <- new.melt[which(new.melt$transcripts==trans),]
print(ggplot(df,aes(x=Group, y=value,fill=Group)) + geom_boxplot() + ylab("PSI") + theme_bw() +
ggtitle(paste(str_wrap(trans, width = 75))) + xlab("") +
theme(axis.text.x = element_text(color = "black",size = 14, angle = 90, hjust = 1,vjust = 0.5),
plot.title = element_text(hjust=0.5,size = 12), legend.position = "none") )
}
dev.off()
## quartz_off_screen
## 2
hbrain <- hbrain[order(hbrain$Group),]
ra <- HeatmapAnnotation(Group = hbrain$Group,col = list(Group=c("Postnatal"= "purple",
"Adult" = "coral3",
"Midstage_Embryo"="blue",
# "Early_Embryo" = "yellow",
"Late_Embryo" = "brown",
"Very_Late_Embryo"="black",
"Adolescent"= "cyan3")))
# "Developing_Embryo"="deeppink1")))
#ha <- HeatmapAnnotation(SpliceType = splice_type, col=list(SpliceType=c("SE"="red","MXE"="blue","RI"="cyan3","A3SS"="yellow","A5SS"="orange")))
lt05 <- rownames(tab)[which(tab$FDR < 0.05)]
toplot <- hbrain.psi.complete[lt05,]
toplot <- medianCtr(toplot) #log2(toplot+1))
draw(Heatmap(as.matrix(toplot), cluster_columns = F, name = "PSI",column_title = "Top DE Transcripts (FDR < 0.05)\nMouse Hindbrain Samples", clustering_distance_columns = "pearson",clustering_method_columns = "average", bottom_annotation = ra, row_names_max_width = max_text_width(rownames(toplot), gp = gpar(fontsize = 3))),heatmap_legend_side = "left", annotation_legend_side = "left")
kable(hbrain) %>% kable_styling()
|
|
Samples
|
DevelopStage
|
Group
|
|
27
|
2224sTS.Mouse.Cerebellum.14.5.Male
|
14.5
|
Midstage_Embryo
|
|
28
|
2228sTS.Mouse.Cerebellum.14.5.Female
|
14.5
|
Midstage_Embryo
|
|
29
|
2238sTS.Mouse.Cerebellum.14.5.Male
|
14.5
|
Midstage_Embryo
|
|
30
|
2244sTS.Mouse.Cerebellum.14.5.Female
|
14.5
|
Midstage_Embryo
|
|
39
|
2814sTS.Mouse.Cerebellum.13.5.Female
|
13.5
|
Midstage_Embryo
|
|
40
|
2822sTS.Mouse.Cerebellum.13.5.Male
|
13.5
|
Midstage_Embryo
|
|
41
|
2830sTS.Mouse.Cerebellum.13.5.Female
|
13.5
|
Midstage_Embryo
|
|
42
|
2838sTS.Mouse.Cerebellum.13.5.Male
|
13.5
|
Midstage_Embryo
|
|
23
|
2186sTS.Mouse.Cerebellum.15.5.Male
|
15.5
|
Late_Embryo
|
|
24
|
2190sTS.Mouse.Cerebellum.15.5.Female
|
15.5
|
Late_Embryo
|
|
25
|
2194sTS.Mouse.Cerebellum.15.5.Male
|
15.5
|
Late_Embryo
|
|
26
|
2218sTS.Mouse.Cerebellum.15.5.Female
|
15.5
|
Late_Embryo
|
|
31
|
2255sTS.Mouse.Cerebellum.16.5.Female
|
16.5
|
Late_Embryo
|
|
32
|
2258sTS.Mouse.Cerebellum.16.5.Male
|
16.5
|
Late_Embryo
|
|
33
|
2274sTS.Mouse.Cerebellum.16.5.Female
|
16.5
|
Late_Embryo
|
|
34
|
2277sTS.Mouse.Cerebellum.16.5.Male
|
16.5
|
Late_Embryo
|
|
19
|
2158sTS.Mouse.Cerebellum.17.5.Male
|
17.5
|
Very_Late_Embryo
|
|
20
|
2161sTS.Mouse.Cerebellum.17.5.Male
|
17.5
|
Very_Late_Embryo
|
|
21
|
2164sTS.Mouse.Cerebellum.17.5.Female
|
17.5
|
Very_Late_Embryo
|
|
22
|
2166sTS.Mouse.Cerebellum.17.5.Female
|
17.5
|
Very_Late_Embryo
|
|
35
|
2308sTS.Mouse.Cerebellum.18.5.Male
|
18.5
|
Very_Late_Embryo
|
|
36
|
2314sTS.Mouse.Cerebellum.18.5.Female
|
18.5
|
Very_Late_Embryo
|
|
37
|
2318sTS.Mouse.Cerebellum.18.5.Male
|
18.5
|
Very_Late_Embryo
|
|
38
|
2324sTS.Mouse.Cerebellum.18.5.Female
|
18.5
|
Very_Late_Embryo
|
|
1
|
1741sTS.Mouse.Cerebellum.0dpb.Female
|
0.0
|
Postnatal
|
|
7
|
1901sTS.Mouse.Cerebellum.0dpb.Female
|
0.0
|
Postnatal
|
|
8
|
1907sTS.Mouse.Cerebellum.0dpb.Male
|
0.0
|
Postnatal
|
|
9
|
1919sTS.Mouse.Cerebellum.3dpb.Female
|
3.0
|
Postnatal
|
|
10
|
1925sTS.Mouse.Cerebellum.3dpb.Male
|
3.0
|
Postnatal
|
|
11
|
1931sTS.Mouse.Cerebellum.3dpb.Female
|
3.0
|
Postnatal
|
|
18
|
1965sTS.Mouse.Cerebellum.3dpb.Male
|
3.0
|
Postnatal
|
|
43
|
3919sTS.Mouse.Cerebellum.0dpb.Male
|
0.0
|
Postnatal
|
|
2
|
1877sTS.Mouse.Cerebellum.2wpb.Male
|
14.0
|
Adolescent
|
|
3
|
1881sTS.Mouse.Cerebellum.2wpb.Male
|
14.0
|
Adolescent
|
|
4
|
1885sTS.Mouse.Cerebellum.2wpb.Female
|
14.0
|
Adolescent
|
|
5
|
1893sTS.Mouse.Cerebellum.4wpb.Male
|
28.0
|
Adult
|
|
6
|
1897sTS.Mouse.Cerebellum.4wpb.Male
|
28.0
|
Adult
|
|
12
|
1937sTS.Mouse.Cerebellum.4wpb.Female
|
28.0
|
Adult
|
|
13
|
1941sTS.Mouse.Cerebellum.4wpb.Female
|
28.0
|
Adult
|
|
14
|
1945sTS.Mouse.Cerebellum.9wpb.Male
|
63.0
|
Adult
|
|
15
|
1949sTS.Mouse.Cerebellum.9wpb.Female
|
63.0
|
Adult
|
|
16
|
1955sTS.Mouse.Cerebellum.9wpb.Male
|
63.0
|
Adult
|
|
17
|
1961sTS.Mouse.Cerebellum.9wpb.Female
|
63.0
|
Adult
|
